Gibbs

Search…
Gibbs
​Robert and Casella (2013) gives the following algorithm,
while Liu, Jun S. (2008) introduces two types of Gibbs sampling strategy.
Bivariate Gibbs sampler
It is easy to implement this sampler:
## Julia program for Bivariate Gibbs sampler
## author: weiya <[email protected]>
## date: 2018-08-22
​
function bigibbs(T, rho)
    x = ones(T+1)
    y = ones(T+1)
    for t = 1:T
        x[t+1] = randn() * sqrt(1-rho^2) + rho*y[t]
        y[t+1] = randn() * sqrt(1-rho^2) + rho*x[t+1]
    end
    return x, y
end
​
## example
bigibbs(100, 0.5)
Completion Gibbs Sampler
Example:
We can use the following Julia program to implement this algorithm.
## Julia program for Truncated normal distribution
## author: weiya <[email protected]>
## date: 2018-08-22
​
# Truncated normal distribution
function rtrunormal(T, mu, sigma, mu_down)
    x = ones(T)
    z = ones(T+1)
    # set initial value of z
    z[1] = rand()
    if mu < mu_down
        z[1] = z[1] * exp(-0.5 * (mu - mu_down)^2 / sigma^2)
    end
    for t = 1:T
        x[t] = rand() * (mu - mu_down + sqrt(-2*sigma^2*log(z[t]))) + mu_down
        z[t+1] = rand() * exp(-(x[t] - mu)^2/(2*sigma^2))
    end
    return(x)
end
​
## example
rtrunormal(1000, 1.0, 1.0, 1.2)
Slice Sampler
​Robert and Casella (2013) introduces the following slice sampler algorithm,
and Liu, Jun S. (2008) also presents the slice sampler with slightly different expression:
In my opinion, we can illustrate this algorithm with one dimensioanl case. Suppose we want to sample from normal distribution (or uniform distribution), we can sample uniformly from the region encolsed by the coordinate axis and the density function, that is a bell shape (or a square).
Consider the normal distribution as an instance.
It is also easy to write the following Julia program.
## Julia program for Slice sampler
## author: weiya <[email protected]>
## date: 2018-08-22
​
function rnorm_slice(T)
    x = ones(T+1)
    w = ones(T+1)
    for t = 1:T
        w[t+1] = rand() * exp(-1.0 * x[t]^2/2)
        x[t+1] = rand() * 2 * sqrt(-2*log(w[t+1])) - sqrt(-2*log(w[t+1]))
    end
    return x[2:end]
end
​
## example
rnorm_slice(100)
Data Augmentation
A special case of Completion Gibbs Sampler.
Let's illustrate the scheme with grouped counting data.
And we can obtain the following algorithm,
But it seems to be not obvious to derive the above algorithm, so I wrote some more details
​Liu, Jun S. (2008) also presents the DA algorithm which based on Bayesian missing data problem.
Then he argues that mmm
 copies of ymis\mathbf y_{mis}ymis​
 in each iteration is not really necessary. And briefly summary the DA algorithm:
It seems to agree with the algorithm presented by Robert and Casella (2013).
Another example:
It seems that we do not need to derive the explicit form of g(x,z)g(x, z)g(x,z)
, if we can directly obtain the conditional distribution. We can use the following Julia program to sample.
## Julia program for Grouped Multinomial Data (Ex. 7.2.3)
## author: weiya <[email protected]>
## date: 2018-08-26
​
# call gamma function
#using SpecialFunctions
​
​
# sample from Dirichlet distributions
using Distributions
​
function gmulti(T, x, a, b, alpha1 = 0.5, alpha2 = 0.5, alpha3 = 0.5)
    z = ones(T+1, size(x, 1)-1) # initial z satisfy `z <= x`
    mu = ones(T+1)
    eta = ones(T+1)
    for t = 1:T
        # sample from g_1(theta | y)
        dir = Dirichlet([z[t, 1] + z[t, 2] + alpha1, z[t, 3] + z[t, 4] + alpha2, x[5] + alpha3])
        sample = rand(dir, 1)
        mu[t+1] = sample[1]
        eta[t+1] = sample[2]
        # sample from g_2(z | x, theta)
        for i = 1:2
            bi = Binomial(x[i], a[i]*mu[t+1]/(a[i]*mu[t+1]+b[i]))
            z[t+1, i] = rand(bi, 1)[1]
        end 
        for i = 3:4
            bi = Binomial(x[i], a[i]*eta[t+1]/(a[i]*eta[t+1]+b[i]))
            z[t+1, i] = rand(bi, 1)[1]
        end
    end
    return mu, eta
end
​
# example
​
## data
a = [0.06, 0.14, 0.11, 0.09];
b = [0.17, 0.24, 0.19, 0.20];
x = [9, 15, 12, 7, 8]; 
​
gmulti(100, x, a, b)
Reversible Data Augmentation
Reversible Gibbs Sampler
Random Sweep Gibbs Sampler
Random Gibbs Sampler
Hybrid Gibbs Samplers
Metropolization of the Gibbs Sampler
Let us illuatrate this algorithm with the following example.
Minimum of Two Exponential
Comparing two groups
Last modified 3yr ago
Copy link
Outline
Bivariate Gibbs sampler
Completion Gibbs Sampler
Slice Sampler
Data Augmentation
Reversible Data Augmentation
Reversible Gibbs Sampler
Random Sweep Gibbs Sampler
Random Gibbs Sampler
Hybrid Gibbs Samplers
Metropolization of the Gibbs Sampler